Introduction Stroke is an important public health issue. It is the second most common cause of death worldwide, after cardiovascular disorders [1] . It is currently the seventh highest cause of disability, and it is predicted that by the year 2020 it will be the fourth highest cause [2] . Because most stroke patients survive beyond the acute period, it is important to reduce the impact of stroke morbidity. Efforts to reduce stroke burden require accurate assessment of outcomes in order to

more »

... e treatments. Morbidity from stroke has many facets. Outcomes reported in stroke trials include impairments, disability, handicap, and quality of life. Each of these may be assessed using a variety of measurement tools. The Rankin Scale (RS) [3] has been used as an outcome measure of stroke trials and routine care for over four decades. Modifi cations to the original scale have been made, resulting in the Modifi ed Rankin Scale (MRS) [4] and the Oxford Handicap Scale (OHS) [5] . In order for an outcome measure to be useful in clinical trials, it must be valid for the purpose that it is intended, reliable and responsive. Some clinimetric properties of the RS [6] and the MRS [7-9] have been discussed previously, however, no comprehensive appraisal of the clinimetric properties of the RS and its derivatives has been published. In this study we performed a critical appraisal of the clinimetric properties of the RS, MRS, and OHS pertaining to their use as outcome measures in stroke trials. We also reviewed the use of these scales in a selection of articles to illustrate concerns raised by our critical appraisal. Abstract Background and Purpose: Efforts to reduce stroke burden require accurate assessment of outcomes in order to compare treatments. The Rankin Scale and its derivatives, the Modifi ed Rankin Scale and the Oxford Handicap Scale, taken together, are among the most common outcome measures that have been used in stroke research. The aim of this study was to perform a critical appraisal of the clinimetric properties of these scales. It was also planned to review the use of these scales in a selection of articles to illustrate concerns raised by the critical appraisal. Summary of Review: A literature search was performed using electronic databases to locate relevant articles about the reviewed scales. The scales were appraised using a structured format regarding the following properties: purpose, development, presentation, language, method of administration, content validity, face validity, feasibility, construct validity, reliability, responsiveness, and generalizability. There are concerns in each of the appraised areas regarding the clinimetric properties of these scales. Conclusion: Further work is needed to improve the clinimetric properties of the reviewed scales to ensure that they are more useful tools in determining the outcome of stroke. Alternatively, a newer global outcome scale with improved clinimetric properties may be a better option for future stroke research.